Genome-wide association studies hold the promise of comprehensive and systematic identification of the genetic basis of natural trait variation. By quantifying how genetic polymorphisms induce large, small, precise, general, or conditional effects on traits, we can unlock a vast reservoir of natural variation that will help us understand how biological systems function and evolve. However, good statistical power to identify variants with low population frequencies or modest effects requires sample sizes that are generally prohibitive. Here, we develop a new approach, "Evolve and Resequence" (E&R), that overcomes this barrier using the genetic model system Drosophila. We have re-purposed experimental evolution of D. melanogaster, which has been used for over 100 years to address fundamental questions in population genetics, and adapted it for use in genome-wide mapping. Together with high-throughput behavioral quantification and various statistical approaches, our technique will be used to build models of behavior based on genotype. The predictive power of these models will tell us a great deal about the nature of genetic information and the molecular systems that translate this information into behavioral output. The long-term goal of this research is to characterize the genetic and environmental influences that result in variation in courtship behavior, for both males and females, in multiple Drosophila species. This work begins with characterization of male courtship song production: during courtship, a Drosophila male extends a single wing and "twangs" this wing repeatedly to produce a wing vibration song consisting of a series of pulses. These traits are among the best characterized of Drosophila behaviors, both in D. melanogaster and across the genus, and they are currently the focus of intensive efforts in neurobiology and molecular genetics. As such, these traits serve as a model system for studying the genetic basis of behavior in animals. Specifically, this proposal aims to 1) locate and characterize the genetic variants affecting courtship song in D. melanogaster, 2) do the same for the closely related D. simulans, to compare and contrast results from multiple species, and 3) begin molecular genetic investigation of the genes and gene systems discovered in aims 1 and 2. PUBLIC HEALTH RELEVANCE: There is tremendous genetic diversity in populations of nearly all species, including humans. In model systems, we can use genetic variation as a tool to understand how heritable information is encoded and deployed, but this requires the development of new experimental and statistical tools. Our proposal develops new innovations in order to tap into this resource in Drosophila to understand the genetic basis of behavior.